پویایی صنعتی افزایش قیمت در دو اقتصاد کوچک باز : آیا سیاست رقابت ملی مهم است؟

In this paper we investigate the impact of competition policy on the level and the dynamics of firm price cost margins in two European countries, Belgium and the Netherlands. We study three questions: (i) Did the changes in the competition law and policy that took place in Belgium in 1993 have an effect on the level and the evolution of the price–cost margins of firms? (ii) Are price mark-ups in the Netherlands, where there was no strict competition policy until recently, higher than in Belgium? (iii) Is openness (import competition) sufficient to discipline firm behaviour? We find that the introduction of a new competition policy in Belgium did not have any effect on the price mark-ups in Belgium. We also find that mark-ups in the Netherlands are higher than those in Belgium. Finally we find that import competition does not lead to lower price cost margins. We offer a number of explanations and implications.

Competitive pressure is generally seen as a good thing in economics, since it reduces monopoly power and forces firms to organise production more efficiently. Welfare increases when markets become more competitive, and hence governments have an interest in establishing and maintaining competitive product markets. As such an antitrust legislation has been implemented in the US since 1890. For more than a century, the Department of Justice and later on the Federal Trade Commission have investigated a countless number of cases, the most recent one being the case against Microsoft.1 On the contrary most European countries adopted competition laws only after World War II. Moreover these were not applied coherently, except perhaps in Germany and in the UK.2 Parallel to these, the founding Treaty of the European Community has included tough antitrust rules against agreements between firms (Art. 85) and against the abuse of a dominant position (Art. 86). Under these rules, most attention has been given to agreements and abuses that had cross-border implications, for example firms that were blocking imports into their country. Within countries, the enforcement of antitrust rules was rather disparate. With the Maastricht Treaty, subsidiarity as a principle was also implemented within this area of policy making. Therefore, countries that were lagging behind have recently brought their legislation closer to European standards. In this paper we investigate the impact of competition policy on the level and the dynamics of firm price cost margins in two European countries, Belgium and the Netherlands. Both are small open economies with very similar economic characteristics. However, in 1993, Belgium adopted a new antitrust legislation very similar to the European one, while the Netherlands continued the implementation of their antitrust law dating from the fifties until January 1998. The effect of competition policy on firm behaviour should be approached from a dynamic perspective, rather than a static one, because the competitive process is itself dynamic. We will focus on the evolution and the level of price mark-ups of firms as suggested by the recent work of Sutton (1991): Equilibrium prices (P) or mark-ups are a declining function of the number of firms (n) in the market, however, the slope may differ depending on the degree of competition in the market. In one extreme case, tacit collusion, P(n) is a flat line, i.e. when a new firm enters the market equilibrium prices are not affected. This is the situation in which firms face very weak price competition. The other extreme case is the one of Bertrand competition where prices fall to marginal costs once a second firm enters the market. This is referred to as the extreme case of very tough price competition. All other oligopoly models will have associated P(n) functions that lie between these two extreme cases. While the strategic interactions between firms can affect the position and the slope of the P(n) function also a number of exogenous parameters, such as the competition law can have an effect on the position of the P(n) function. In this sense, competition policy could lead to tougher price competition, which may in fact lead to less entry in the market because unit margins are reduced in case the competition policy focuses on the level of the margins. 3 While we will not study the implications for firm entry or exit, we will study the impact of competition policy on price mark-ups of firms. This paper makes various contributions. First, we test whether competition policy has an effect on this P(n) function or on the mark-up of firms. We test this in two ways: First, we test whether a change in the competition policy in the case of Belgium has led to a decrease in the price mark-ups. Most other papers in the literature have focused on the static behaviour of firms. In particular, most papers have used time series to estimate a mark-up, which is assumed to be constant over time, and hence it is assumed that competition that firms face is rather static in nature. Here we want to look at the dynamic pricing behaviour after a change in competition policy took place. Second, we test whether the presence of competition policy leads on average to lower mark-ups by comparing Belgium which has competition policy with the Netherlands where there was no competition policy until recently. By studying the dynamics of price cost margins we hope to gain insights into the effectiveness and role of competition policy. The second contribution of this paper lies in the data and the methodology that we use. We have an exceptionally rich firm level panel data set of Belgian and Dutch firms covering the years 1992–1996. Most studies have used aggregate data or sector level data (Domowitz et al., 1988 and Hall, 1988, among others). In contrast, the use of firm level data not only increases the reliability and efficiency of the estimates due to an increased number of observations, but also allows us to take into account firm-heterogeneity within sectors. In addition, the use of firm level data allows us to construct good instruments and allows us to compute mark-ups at the sector level. Furthermore, as suggested by Levinsohn (1993) the estimating equation is derived from the theory of the firm and therefore should require firm level data. The fact that we use firm level panel data also allows us to study the dynamics of price cost margins, which is not possible if only a single cross-section is used. In addition to testing the effects of competition policy on mark-ups, we also compare the level of mark-ups across different sectors and test whether there are other mechanisms that may discipline firms. In particular, both Belgium and the Netherlands are small open economics, so both countries are vulnerable to competition from imports, which potentially has an impact on price–cost margins. We explore whether import competition disciplines firm behaviour. This allows us to look at the question of the complementarity between competition policy and trade policy. The remainder of this paper is organised as follows. In the next section we present the policy changes in both Belgium and the Netherlands. We then discuss the econometric methodology in Section 3. In Section 4, we describe the data and report the results. Section 5 concludes.

In this paper we analysed mark-ups in the Belgian and Dutch manufacturing industries. This paper made various contributions to the literature. First, while most papers use data at the sector or aggregate level, we used a unique firm level panel data set covering several years. This increases the reliability of the estimates, it allowed us to construct valid instruments and to study the dynamic behaviour of price mark-ups. Second, we estimated whether different regimes of competition policy affected the mark-up of firms in a number of ways: (i) we tested whether we can learn anything by analysing the dynamics of competition policy in relation to the price–cost margins. For the Belgian case we had a unique event where we could test whether the new competition law that was introduced led to a decline in the price mark-up, which we compared with the Netherlands where there was no new competition law for the same period; (ii) we tested whether the average mark-up in the Netherlands, where there was no up-to-date competition law until recently, was higher than in Belgium, characterised by a stringent competition law since 1993, (iii) we tested whether in small open economies like Belgium and the Netherlands import competition disciplined firm behaviour. Our results indicate that competition policy in Belgium did not have an effect on price–cost margins, which suggests that the old price regulatory system already disciplined firms in a substantial way. Together with strong import competition, this turned the role of antitrust policy in Belgium into a very modest one. This interpretation is strengthened by the fact that the mark-up in the Netherlands is always estimated higher than the ones in Belgium, even if we do the estimation at the two-digit sector level. This, however, does not imply the redundancy of such a policy: the price controls being abolished and the role of import competition not guaranteed, as illustrated in the Netherlands, a continuous effort to monitor the evolution of price–cost margins is needed. Finally, we find the surprising result that sectors with high import penetration in the Netherlands have higher price mark-ups than sectors with low import penetration; while in Belgium there is no statistically significant difference between the type of sectors. This suggests that competition policy and openness should not be seen as substitutes, but rather as complements. This also suggests again that competition policy and trade policy are closely related.